PHENOMENA AND PRINCIPLES OF BREEDING 469 



of the direct progeny of that individual would be likely to inherit 

 that character. If, then, two of the offspring possessing the char- 

 acter were bred together, the chances of its appearance in their 

 offspring would be one fourth from each parent and one sixteenth 

 from the grandparent. Nine in every sixteen of the second gen- 

 eration would inherit the character. As by constant selection the 

 number of ancestors which had the character is increased, and the 

 proportion of ancestors which did not have it is steadily reduced 

 and its influence rapidly diminished, only a few generations are 

 required to reach the stage of fixity of the character in the race 

 where the influence of ancestors unlike in respect to it becomes 

 a negligible factor. 



Galton's law is not a law or rule of practice in poultry breeding. 

 The attitude of the practical poultry breeder toward it should not 

 be misunderstood ; it cannot be said that he uses it. As a formal 

 statement based on scientific investigation it has been especially 

 serviceable to those giving instruction in the principles of breed- 

 ing, to prove the general rule of selection, to demonstrate the 

 stability and practical purity of new breeds and varieties, and to 

 show the need of close breeding to fix and hold desired combi- 

 nations of characters. 



Mendel's law. Of more importance than Galton's statement 

 were the discoveries of Mendel in regard to the behavior of unlike 

 characters in transmission. When first published by Mendel, these 

 attracted no attention. Mendel's account of his work was redis- 

 covered about 1900, and has since profoundly influenced the course 

 of investigation of the subject of heredity. Unfortunately many 

 scientists who took up this work with enthusiasm failed to note 

 some serious faults in Mendel's treatment of his results and in his 

 enunciation of principles based upon them, and consequently, though 

 a considerable amount of this work has been done with poultry, it 

 has not yet yielded results of such value to poultry breeders as at 

 first seemed likely to follow scientific investigation in this field. 



Mendel, experimenting mostly with the sweet pea, observed 1 

 (i) that in the offspring of certain crosses a certain character of a 

 parent form might disappear ; (2) that when these offspring were 



' For a fuller statement of Mendel's law see Davenport's " Principles of 

 Breeding." 



